US4920540A - Fault-tolerant digital timing apparatus and method - Google Patents

Fault-tolerant digital timing apparatus and method Download PDF

Info

Publication number
US4920540A
US4920540A US07018629 US1862987A US4920540A US 4920540 A US4920540 A US 4920540A US 07018629 US07018629 US 07018629 US 1862987 A US1862987 A US 1862987A US 4920540 A US4920540 A US 4920540A
Authority
US
Grant status
Grant
Patent type
Prior art keywords
clock
logic
signal
output
input
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
US07018629
Inventor
Kurt F. Baty
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Stratus Computer Inc
Original Assignee
Stratus Computer Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Grant date

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/07Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/16Error detection or correction of the data by redundancy in hardware
    • G06F11/1604Error detection or correction of the data by redundancy in hardware where the fault affects the clock signals of a processing unit and the redundancy is at or within the level of clock signal generation hardware
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/07Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/14Error detection or correction of the data by redundancy in operation
    • G06F11/1402Saving, restoring, recovering or retrying
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/07Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/16Error detection or correction of the data by redundancy in hardware
    • G06F11/1608Error detection by comparing the output signals of redundant hardware
    • G06F11/1625Error detection by comparing the output signals of redundant hardware in communications, e.g. transmission, interfaces
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/07Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/16Error detection or correction of the data by redundancy in hardware
    • G06F11/20Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
    • G06F11/2002Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where interconnections or communication control functionality are redundant
    • G06F11/2005Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where interconnections or communication control functionality are redundant using redundant communication controllers
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/07Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/16Error detection or correction of the data by redundancy in hardware
    • G06F11/20Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
    • G06F11/2002Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where interconnections or communication control functionality are redundant
    • G06F11/2007Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where interconnections or communication control functionality are redundant using redundant communication media
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/07Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/16Error detection or correction of the data by redundancy in hardware
    • G06F11/20Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
    • G06F11/2017Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements where memory access, memory control or I/O control functionality is redundant
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/22Detection or location of defective computer hardware by testing during standby operation or during idle time, e.g. start-up testing
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F1/00Details not covered by groups G06F3/00 – G06F13/00 and G06F21/00
    • G06F1/04Generating or distributing clock signals or signals derived directly therefrom
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/07Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/16Error detection or correction of the data by redundancy in hardware
    • G06F11/1629Error detection by comparing the output of redundant processing systems
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/07Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/16Error detection or correction of the data by redundancy in hardware
    • G06F11/1666Error detection or correction of the data by redundancy in hardware where the redundant component is memory or memory area
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/07Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/16Error detection or correction of the data by redundancy in hardware
    • G06F11/20Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements

Abstract

Computer timing apparatus enables two clock elements to produce a single stream of timing pulses, without interruption, when both elements are operating normally, and when one element fails. The apparatus incorporates a multi-stable stage and an output logic stage. The multi-stable stage detects state transitions in the input signals of each clock element and generates a corresponding clock-tracking signal which can disable the output of the corresponding clock from propagating through the output logic. The output logic stage logically combines each clock signal with its corresponding clock-tracking signal, and logically combines the resultant signal to produce a single stream of output signals responsive to a next transition produced by either of the two clock elements.

Description

BACKGROUND OF THE INVENTION

This invention relates to computer timing apparatus and, more particularly, to apparatus and methods for providing highly reliable clock signals for operating digital data processing equipment and systems.

Digital computer equipment commonly includes a clock device to produce timing pulses for synchronizing and sequencing operations. This invention provides such a clock device that operates without interruption in the event of certain faults.

Fault conditions are inevitable in digital computer systems, due in part to the number and complexity of components and circuits they employ. Computers have included redundant processor modules and redundant memory modules, for example, to continue operation in the event of module failure.

Similarly, fault conditions can occur in the digital clock devices which control the timing of digital computer equipment. Computer equipment employing prior art clock apparatus can become disabled by a single clock fault.

It has proven difficult, however, to provide redundancy for clock modules. It is accordingly an object of this invention to provide a clock apparatus and method which operates with improved tolerance to faults and hence with improved reliability.

It is a further object of the invention to provide a clock device having two clock elements, and which provides an uninterrupted stream of output clock pulses notwithstanding failure of either clock element.

Other general and specific objects of the invention will in part be obvious and will in part appear hereinafter.

SUMMARY OF THE INVENTION

Computer timing apparatus according to the invention enables two redundant clock elements to produce a single stream of timing pulses, without interruption, when both elements are operating normally and when one element fails so that only the other one is operating normally. Where even greater reliability is desired, the invention can be practiced with more than two clock elements. In such an expanded embodiment, a reliable stream of timing pulses is produced so long as any one clock element operates in each timing interval.

More particularly, computer clock apparatus according to the invention has at least first and second clock elements for producing respective first and second streams of clock input pulses. A comparator element compares pulses produced from the two clock elements, and detects state transitions in each of the two pulse streams. The apparatus further includes an output element responsive to the comparator element and in communication with the clock elements. The output element responds to the detection of a state transition to produce a clock output pulse representative of a next clock pulse produced by either clock element. As a result, the apparatus generates an uninterrupted clock output signal notwithstanding failure of any single clock element.

The invention thus makes it possible to increase the reliability against failure of clock elements, by the expedient of providing a logic circuit that combines the output pulse streams from redundant clock elements and responds to any one clock element, with substantially uninterrupted output timing.

The invention comprises steps and apparatus embodying features of construction, combinations of elements and arrangements of parts adapted to effect such steps, as exemplified in the following detailed disclosure, and the scope of the invention is indicated in the claims.

BRIEF DESCRIPTION OF THE DRAWINGS

For a fuller understanding of the nature and objects of the invention, reference should be made to the following detailed description and the accompanying drawings, in which:

FIG. 1 is an electrical block diagram of a two-clock system in accordance with the invention;

FIG. 2 is a schematic diagram of a preferred logic circuit for the system of FIG. 1; and

FIG. 3 is a timing diagrams illustrating operation of the two-clock system of FIG. 2.

DESCRIPTION OF ILLUSTRATED EMBODIMENT

The invention, in one aspect, provides a circuit for testing plural timing signals, referred to as "strobe" signals, which may have different frequencies and duty cycles, and for generating a single output strobe signal from them. The invention will be described in connection with a reliable clock embodiment. However, it will be apparent to one skilled in the art that the invention can be embodied in a variety of structures and systems for testing and generating strobe signals.

A digital clock device 10 according to the invention has, as FIG. 1 shows, a first clock element 11, a second clock element 12, and a clock logic element 15. The clock logic element 15 receives an input stream of timing signals, referred to as CLK1, from the first clock element 11 over line 13, and receives a CLK2 input stream of timing signals from the second clock element 12 over line 14. Clock logic element 15 compares the CLK1 and CLK2 signals, in a manner more fully discussed hereinafter, and produces a stream of output clock Pulses, designated CLK*, which remains substantially uninterrupted notwithstanding failure of either clock element. The output clock pulse stream CLK*, carried on line 16, may be used to clock digital computer equipment such as a central processor unit 17 as disclosed in the above-mentioned U.S. Pat. No. 4,453,215.

FIG. 2 shows a logic circuit implementation of the two-clock system 10 of FIG. 1. The clock elements 11 and 12 normally generate streams of clock signals CLK1 and CLK2, respectively, carried on lines 13 and 14, respectively. Clock elements 11 and 12 may be constructed with oscillator circuits known in the art. The respective output signals CLK1 and CLK2 of clock elements 11 and 12 will therefore preferably consist of alternating high and low digital values. In a preferred embodiment of the invention, clock elements 11 and 12 are synchronized with one another and otherwise operate independently, and have identical frequency and hence identical clock rates. However, clock elements 11 and 12 can have different frequencies and different duty cycles, although, as discussed below, indeterminate operation may occur when one clock element produces a rising edge coincidentally or simultaneously with production of a falling edge by the other.

The circuit of FIG. 2 implements clock logic element 15 of FIG. 1 with an input section formed by NAND gates 21 and 22, an intermediate multi-stable section that employs two flip flops 23 and 24, and an output section that employs two AND gates 25 and 26 and a NOR gate 27. More particularly, NAND gate 21 is configured as an inverter, as known in the art. Both inputs of NAND gate 21 are driven by pulse stream CLK1 over line 13, and the output of NAND gate 21 is a CLK1* signal on line 28. The CLK1* signal is inverted and delayed with respect to the CLK1 signal. The delay is attributable to propagation delays through the NAND gate 21.

Similarly, NAND gate 22 inverts the CLK2 signal carried on line 14. The output CLK2* signal of NAND gate 22 is inverted and delayed with respect to the CLK2 signal, and is carried on line 29. The Propagation delay in NAND gate 22 preferably is nominally identical to the delay in NAND gate 21.

The illustrated flip-flops 23 and 24 are D-type flip-flops known in the art. The D input of flip-flop 23 is driven by the CLK2 signal over line 14a. The clock input of flip-flop 23 is driven by signal CLK1* over line 28, and flip-flop 23 is cleared by signal CLK2 over line 14b. The signal at the inverting Q output of flip-flop 23, referred to as TRACK2, is carried over line 210.

Similarly, the D input of flip-flop 24 is driven by signal CLK1 over line 13b, the clock input of flip-flop 24 is driven by signal CLK2* over line 29, and the flip-flop is cleared by signal CLK1 over line 13c. The signal at the inverting Q output of flip-flop 24, referred to as signal TRACK1, is carried over line 211.

AND gate 25 performs an AND operation with the TRACK2 and CLK2 signals, carried over lines 210 and 14brespectively. The output of AND gate 25, referred to as signal AND2, is carried over line 212. AND gate 26 similarly performs an AND operation with the TRACK1 and CLK1 signals, carried over lines 211 and 13a, respectively. The output of AND gate 26, referred to as signal AND1, is carried over line 213.

NOR gate 27 performs a NOR operation with the AND1 and AND2 signals. The output signal from the NOR gate 27, referred to as signal CLK*, is the output of the two-clock device and is carried over line 16. Flip flops 23 and 24, AND gates 25 and 26, and NOR gate 27 operate with nominally identical propagation delays. The nominal propagation delays of the components are such that the CLK1 and CLK2 signals normally propagate through the AND and NOR gates of the output section with a cumulative delay less than that presented by the inverter and flip flop stages. In order to ensure proper circuit function, the total propagation delay of NOR gate 27 and of either AND gate 25 or 26 is to be less than the total propagation delay of inverter 21 and flip flop 23, and of inverter 22 and flip flop 24

In operation, the circuit of FIG. 2 generates a rising edge in signal CLK* whenever both signal CLK1 and signal CLK2 fall. Conversely, the circuit generates a falling edge in signal CLK* whenever both signal CLK1 and signal CLK2 rise. If one clock element fails to generate a rising or falling edge, the circuit generates a state transition at the CLK* output in response to the next rising or falling edge produced by the remaining operational clock element. The output in such a case may be delayed with respect to the input. The delay is attributable to propagation time through the flip-flops. The frequency of the resultant CLK* signal thus is invariant, so long as one clock element produces a state transition in the proper time interval. The exact transition times of the CLK* signal may exhibit minor propagation delays, due to the operating conditions of the clock elements, and the frequency is substantially stable.

The multi-stable stage consisting of flip flops 23 and 24 detects upward and downward state transitions in the clock output signals CLK1 and CLK2, and disables the output of a failed clock element from propagating through the AND gates 25 and 26 of the output stage. The operation of the multi-stable section is described in greater detail below in connection with Table I.

Table I summarizes the operation of the two-clock system of FIG. 2 for eight successive timing intervals, commencing at times t0, t1, t2, . . . t7. Table I illustrates a sequence in which the two clock elements operate properly except in Intervals Three and Seven.

                                  TABLE I__________________________________________________________________________                    AND1  AND2  CLK* -     [= CLK1 [= CLK2 [= AND1                    AND   AND   NORIntervalCLK1    CLK2        TRACK 1              TRACK 2                    TRACK 1]                          TRACK 2]                                AND2]__________________________________________________________________________0    0   0   1     1     0     0     11    1   1   1     1     1     1     02    0   0   1     1     0     0     13    1   0   1     1     1     0     04    0   0   1     1     0     0     15    1   1   1     1     1     1     06    1   0   0     1     0     0     17    1   1   0     1     0     1     0__________________________________________________________________________

Intervals Zero, One and Two of Table I represent normal operation of the two clock elements 11 and 12 as do intervals Four and Five, as discussed below.

Examination of Table I in conjunction with FIG. 2 shows that the output signal CLK* is a function of the digital values of signals AND1 and AND2, and in turn, that signals AND1 and AND2 result from AND operations performed on signals CLK1 and TRACK1, and CLK2 and TRACK2, respectively. The operation of the circuit of FIG. 2 is thus best described in terms of the values of the AND, TRACK, and CLK signals at each interval shown in Table I.

At interval Zero, clock elements 11 and 12 generate signals CLK1 and CLK2 having a value of logic ZERO. A logic ZERO is therefore asserted at the CLEAR inputs of both flip-flops 23 and 24. Asserting a logic ZERO at the CLEAR input of a D-type flip-flop forces an output of logic ONE at the inverting Q output. Hence at interval Zero, signals TRACK1 and TRACK2, which are generated at the inverting Q outputs of flip-flops 24 and 23, respectively, both have a value of logic ONE.

The signal AND1, which is generated by the AND operation performed by AND gate 26 on signals CLK1 and TRACK1, is a logic ZERO unless both CLK1 and TRACK1 are logic ONE. Similarly, signal AND2, which is generated by the AND operation performed by AND gate 25 on signals CLK2 and TRACK2, is logic ZERO unless both CLK2 and TRACK2 are logic ONE. Thus, in interval Zero, because both TRACK1 and TRACK2 are logic ZERO, the AND1 and AND2 signals each have a value of logic ZERO.

The output signal CLK*, which is the result of the NOR operation performed by NOR gate 27 on signals AND1 and AND2, is a logic ZERO unless both AND1 and AND2 are logic ZERO. As described above, signals AND1 and AND2 have a value of logic ZERO in interval Zero, and accordingly, output signal CLK* is logic ONE in interval Zero.

In interval One, clock elements 11 and 12 continue to operate normally, each rising to logic ONE. Because logic ONEs are thereby asserted at the CLEAR inputs of the D-type flip-flops 23 and 24, the flip-flops are enabled for clocked operation. Flip-flops 23 and 24 are clocked by signals CLK1* and CLK2* respectively, which are inverted with respect to CLK1 and CLK2, respectively. Because a rising edge is generated in signals CLK1 and CLK2 in interval One, a falling edge is asserted at the clock inputs of flip-flops 23 and 24 in interval One. Flip-flops 23 and 24 are, as indicated in schematic form by FIG. 2, rising edge flip-flops, which change state only when signals CLK1* and CLK2*, respectively, rise from logic ZERO to logic ONE. Accordingly, in interval One, flip-flops 23 and 24 do not change state, and signals TRACK1 and TRACK2 each remain at logic ONE.

Because signals CLK1, CLK2, TRACK1 and TRACK2 all are logic ONE during interval One, signals AND1 and AND2 are logic ONE, and signal CLK* is logic ZERO during this interval.

During interval Two, clock elements 11 and 12 continue to operate normally, each falling to logic ZERO. Because logic ZEROs are therefore asserted at the CLEAR inputs of flip-flops 23 and 24, signals TRACK1 and TRACK2 are forced to logic ONE, as discussed above. Because signals CLK1 and CLK2 fall to logic ZERO, signals AND1 and AND2 accordingly fall to logic ZERO, and signal CLK* rises to logic ONE.

In Interval Three, signal CLK2 displays a fault condition in that it remains at logic ZERO, rather than rising to logic One.

A logic ZERO is thus asserted at the CLEAR input of flip-flop 23, forcing output signal TRACK2 to logic ONE. Because signal CLK1 is at logic ONE, a logic ONE is asserted at the CLEAR input of flip-flop 24, setting flip-flop 24 for clocked operation. However, signal CLK2*, which is inverted with respect to signal CLK2, was logic ONE in interval Two, and continues to be at logic ONE in interval Three. Because flip-flop 24 only changes state when a rising edge is applied to its clock input, flip-flop 24 does not change state, and signal TRACK1 continues to have a value of logic ONE.

Accordingly, signal AND1, which is the result of the AND operation performed by gate 26 on signals CLK1 and TRACK1, is logic ONE during interval Three. Signal AND2, which is the result of the AND operation performed by gate 25 on signals CLK2 and TRACK2, is logic ZERO during this interval. Output signal CLK*, which is the result of the NOR operation performed by gate 27 on signals AND1 and AND2, therefore switches to logic ZERO during this interval.

Intervals Four and Five again illustrate normal operation of the two clock elements. In Interval Six of Table I, however, clock element 11 fails to generate a logic ZERO and the CLK1 signal remains at logic ONE.

Because signal CLK1 has a value of logic ONE, a logic ONE is asserted at the CLEAR input of flip-flop 24, which enables flip-flop 24 for clocked operation. CLK2*, which is inverted with respect to CLK2, rises to logic ONE when CLK2 falls to logic ZERO. When the CLK2* rising edge is asserted at the clock input of flip-flop 24, signal TRACK1 at the inverting Q output of flip-flop 24 becomes the complement of what the value at the D input of flip-flop 24 was before the rising edge was asserted. CLK1, at the D input, is a logic ONE during this prior interval, and TRACK1 accordingly falls to logic ZERO during interval Six.

Signal TRACK2 remains at logic ONE during interval Six, because the logic ZERO in signal CLK2, which is asserted at the CLEAR input of flip-flop 23, forces a logic ONE at the inverting Q output of flip-flop 23.

The signal AND1, which is the output of the AND operation which gate 26 performs with TRACK 1 and CLK1 signals, falls to logic ZERO. Signal AND2 also falls to logic ZERO. The CLK* signal output from the NOR gate 27 follows the falling edge of CLK2 and rises to a logic ONE, thereby maintaining an uninterrupted stream of clock pulses during interval Six.

Interval Six in Table I shows that, because AND gate 26 of the output stage will only transmit a logic ONE if both TRACK1 and CLK1 are logic ONE, the TRACK1 signal generated by flip-flop 24 disables a logic ONE output of clock element 11 from propagating through AND gate 26 when clock element 11 fails to properly generate a logic ZERO.

During interval Seven, signal CLK1 remains at logic ONE, and signal CLK2 rises to logic ONE. TRACK1 remains at logic ZERO, because while flip-flop 24 is set for clocked operation by the logic ONE value of signal CLK1, signal CLK2*, which clocks flip-flop 24, falls from logic ONE to logic ZERO, and flip-flop 24 therefore does not change state. Signal TRACK2 remains at logic ONE, because while flip-flop 23 is set for clocked operation by the logic ONE value of signal CLK2, signal CLK1*, which clocks flip-flop 23, remains at logic ZERO. Flip-flop 23 therefore does not change state.

FIG. 3 shows a set of digital timing diagrams for the CLK1, CLK2 and CLK* signals during the timing intervals shown in Table I. The CLK1 wave form in FIG. 3 shows, for example, that at time t3, corresponding to the start of Interval Three of Table I, clock element 12 fails to generate a rising edge. Clock element 11, which continues to operate, generates a rising edge. Signal TRACK1 remains at logic ONE, because although flip-flop 24 is set for clocked operation by the logic ONE of signal CLK1, signal CLK2*, which clocks flip-flop 24, remains at logic ONE. Flip-flop 24 hence does not change state. Signal TRACK2 is forced to logic ONE by the logic ZERO of signal CLK2, which is applied to the CLEAR input of flip-flop 23. Accordingly, signals AND1 and AND2 are logic ONE and ZERO, respectively, during this interval. CLK*, the result of a NOR operation on AND1 and AND2, is thus switched to logic ZERO.

Similarly, at time t5, after transmitting three logic ONES, clock element 11 fails to fall to logic ZERO. Clock element 12, which continues to operate, generates a falling edge. The logic ZERO in CLK2 forces signal TRACK1 to logic ONE, while signal CLK2* clocks flip-flop 24, generating a logic ZERO in signal TRACK1. The logic ZERO in signal TRACK1 forces signal AND1 to logic ZERO, and the logic ZERO in signal CLK2 forces signal AND2 to logic ZERO. Thus, output signal CLK*, which is the result of a NOR operation on signals AND1 and AND2, rises to logic ONE.

The delay, designated "d," which is shown in exaggerated form at time t6, represents propagation delays characteristic of the flip-flop TTL stages. No such delay is indicated in signal CLK* during the fault condition at time t3, because at time t3, the rising edge in signal CLK1 triggers the falling edge in signal CLK* directly through gates 26 and 27.

In a preferred embodiment of the invention, the two input clock elements 11 and 12 normally operate in lock-step synchronism. This normal operating mode is illustrated in Table I at Intervals Zero-Three and Four and Five. Those skilled in the art will understand that the two clocks can be operated substantially in phase by simply starting the two equal-frequency clocks simultaneously. One alternative is to use an analog circuit known in the art to synchronize the clock pulses. Both methods of operating two clocks elements in phase are known in the art, and are within the ambit of the invention.

The examples of Table I and FIG. 3 demonstrate that the illustrated circuit maintains an uninterrupted output of timing pulses as long as one input strobe or clock element continues to operate, by switching in each timing interval. The configuration of gates 21, 22, 25-27 and flip flops 23 and 24 illustrated in FIG. 2 provides a logic section which compares input pulses from plural strobe or clock elements and generates a stream of output pulses. Each output pulse is representative of the next input pulse transition produced by any of the plural input clock elements.

One feature of the invention is that so long as the clock elements operate with timing phases which are within 180 degrees of each other, the output signal from the device which the invention provides switches from ONE to ZERO in response to all inputs rising. Further, if all inputs do not rise, the output signal of the device switches from ONE to ZERO in response to the first rising edge which occurs at the input stage. The output signal switches from ONE to ZERO in response to all inputs falling or, if all inputs do not fall, in response to the first falling edge which occurs at the input stage.

The clock device is subject to fault only if the input clock elements are at least 180 degrees out of phase, for then the logic circuitry may be unable to provide an uninterrupted output of regular clock pulses. When the clock elements are 180 degrees out of phase, a rising edge and a falling edge are simultaneously asserted at the input stage of the circuit. A race condition results, and the output of the circuit is indeterminate, depending on small delays within the gates.

The illustrated circuit can be used with strobe or clock inputs having different duty cycles and different frequencies. The circuit produces a reliable output pulse stream, as long as the strobe or clock inputs do not simultaneously generate a rising and a falling edge. Such a condition of coincident opposite input transitions may occur, for example, if the strobe or clock input devices are in a "beat" condition, in which the frequency of one input device differs slightly from the frequency of another input device. The input devices then "race" each other until one device generates a rising edge and the other device simultaneously generates a falling edge.

Subject to this constraint of avoiding simultaneous assertion of rising and falling edges at the input of the circuit, the circuit need not be driven with "clock" inputs having exacting control of frequency and duty cycle, but may instead be driven by "strobe" inputs, having less regular pulse outputs.

In a preferred embodiment of the invention, logic gates 25, 26 and 27 are provided by a single component known as an AND OR INVERT (AOI) device. An AOI is preferred over three separate gate devices because the single AOI increases switching speed and reliability.

The illustrated implementation of the circuit can be utilized with pulse widths as low as 15 nanoseconds, with 60-100 nanosecond pulse widths and 4 megahertz frequencies being typical.

Additionally, the illustrated circuit can be expanded to use three strobe or clock inputs. Each strobe or clock input is checked by a multi-stable stage, and its output is enabled or disabled from propagating through an AOI logic output stage, as described above in connection with the illustrated circuit. Such a plural clock circuit according to the invention may bear passing resemblance to prior art "voting" circuits which provide an output signal based on the values at a majority of plural inputs. However, prior art voting circuits do not generate an uninterrupted timing pulse output when only a single input strobe element is functioning.

Moreover, the Boolean complement of the illustrated circuit, in which flip-flops 23 and 24 are triggered by falling edges rather than rising edges, will be obvious to those skilled in the art.

It will thus be seen that the invention efficiently attains the objects set forth above, among those made apparent from the preceding description. It will be understood that changes may be made in the above construction and in the foregoing sequences of operation without departing from the scope of the invention. It is accordingly intended that all matter contained in the above description or shown in the accompanying drawings be interpreted as illustrative rather than in a limiting sense.

It is also to be understood that the following claims are intended to cover all of the generic and specific features of the invention as described herein, and all statements of the scope of the invention which, as a matter of language, might be said to fall therebetween.

Claims (2)

Having described the invention, what is claimed as new and secured by Letters Patent is:
1. A digital logic circuit for generating a reliable serial stream of clock output pulses from first and second digital input clocks, said circuit comprising:
A. first and second input means for receiving clock input pulses produced from said first and second clocks respectively,
B. first bi-stable logic means associated with said first input means, for storing a first digital value received at said first input means and selectively transmitting, responsive to pulses received at said second input means, said first digital value or a complementary digital value of said first digital value,
C. second bi-stable logic means associated with said second input means, for storing a second digital value received at said second input means and selectively transmitting, responsive to pulses received at said first input means, said second digital value or a complementary digital value of said second digital value,
D. output logic means, connected with said first and second input means and with said first and second bi-stable logic means, for (i) comparing pulses from said first and second input means, and from said first and second bi-stable logic means, and (ii) producing a clock output pulse in response to a next clock input pulse received at either of said input means, so that an uninterrupted clock pulse output is produced notwithstanding failure of any single one of said input clocks.
2. Computer timing apparatus for producing a single stream of timing pulses continually when at least any one of first and second input sequences of timing signals are present, said apparatus comprising
A. multi-stable binary logic means for receiving said first and second input sequences of timing signals and for producing, in response thereto, first and second sequences of clock-tracking signals, each sequence of clock-tracking signals being responsive to occurrence of a transition in at least one input sequence of timing signals, and
B. output logic means for logically combining each input sequence of timing signals with its corresponding clock-tracking signal to generate a corresponding resultant signal, and for logically combining the resultant signals to produce a single stream of output signals responsive to a next transition in any of said first and second input sequences of timing signals.
US07018629 1987-02-25 1987-02-25 Fault-tolerant digital timing apparatus and method Expired - Fee Related US4920540A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US07018629 US4920540A (en) 1987-02-25 1987-02-25 Fault-tolerant digital timing apparatus and method

Applications Claiming Priority (10)

Application Number Priority Date Filing Date Title
US07018629 US4920540A (en) 1987-02-25 1987-02-25 Fault-tolerant digital timing apparatus and method
US07079223 US4939643A (en) 1981-10-01 1987-07-29 Fault tolerant digital data processor with improved bus protocol
US07079297 US4926315A (en) 1981-10-01 1987-07-29 Digital data processor with fault tolerant peripheral bus communications
US07079218 US4931922A (en) 1981-10-01 1987-07-29 Method and apparatus for monitoring peripheral device communications
DE19883853734 DE3853734T2 (en) 1987-02-25 1988-02-23 Device for fault-tolerant Digitaltaktierung.
DE19883853734 DE3853734D1 (en) 1987-02-25 1988-02-23 Device for fault-tolerant Digitaltaktierung.
EP19880102650 EP0280258B1 (en) 1987-02-25 1988-02-23 Fault-tolerant digital timing apparatus
JP3976488A JPS63296118A (en) 1987-02-25 1988-02-24 Disturbance allowable digital timing apparatus and method
US07368125 US4974150A (en) 1981-10-01 1989-06-16 Fault tolerant digital data processor with improved input/output controller
US07368124 US4974144A (en) 1981-10-01 1989-06-16 Digital data processor with fault-tolerant peripheral interface

Related Child Applications (5)

Application Number Title Priority Date Filing Date
US07079223 Continuation-In-Part US4939643A (en) 1981-10-01 1987-07-29 Fault tolerant digital data processor with improved bus protocol
US07079297 Continuation-In-Part US4926315A (en) 1981-10-01 1987-07-29 Digital data processor with fault tolerant peripheral bus communications
US07079218 Continuation-In-Part US4931922A (en) 1981-10-01 1987-07-29 Method and apparatus for monitoring peripheral device communications
US07368125 Continuation-In-Part US4974150A (en) 1981-10-01 1989-06-16 Fault tolerant digital data processor with improved input/output controller
US07368124 Continuation-In-Part US4974144A (en) 1981-10-01 1989-06-16 Digital data processor with fault-tolerant peripheral interface

Publications (1)

Publication Number Publication Date
US4920540A true US4920540A (en) 1990-04-24

Family

ID=21788941

Family Applications (1)

Application Number Title Priority Date Filing Date
US07018629 Expired - Fee Related US4920540A (en) 1987-02-25 1987-02-25 Fault-tolerant digital timing apparatus and method

Country Status (4)

Country Link
US (1) US4920540A (en)
EP (1) EP0280258B1 (en)
JP (1) JPS63296118A (en)
DE (2) DE3853734D1 (en)

Cited By (25)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5020024A (en) * 1987-01-16 1991-05-28 Stratus Computer, Inc. Method and apparatus for detecting selected absence of digital logic synchronism
US5077739A (en) * 1989-05-17 1991-12-31 Unisys Corporation Method for isolating failures of clear signals in instruction processors
US5220668A (en) * 1990-09-21 1993-06-15 Stratus Computer, Inc. Digital data processor with maintenance and diagnostic system
US5257383A (en) * 1991-08-12 1993-10-26 Stratus Computer, Inc. Programmable interrupt priority encoder method and apparatus
US5313476A (en) * 1991-06-28 1994-05-17 International Business Machines Corporation Clock security ring
US5377206A (en) * 1993-02-03 1994-12-27 Honeywell Inc. Multiple-channel fault-tolerant clock system
US5379381A (en) * 1991-08-12 1995-01-03 Stratus Computer, Inc. System using separate transfer circuits for performing different transfer operations respectively and scanning I/O devices status upon absence of both operations
US5379415A (en) * 1992-09-29 1995-01-03 Zitel Corporation Fault tolerant memory system
US5422896A (en) * 1993-02-24 1995-06-06 Nec Corporation Timing check circuit for a functional macro
US5440751A (en) * 1991-06-21 1995-08-08 Compaq Computer Corp. Burst data transfer to single cycle data transfer conversion and strobe signal conversion
US5471488A (en) * 1994-04-05 1995-11-28 International Business Machines Corporation Clock fault detection circuit
US5588111A (en) * 1988-12-09 1996-12-24 Tandem Computers, Incorporated Fault-tolerant computer system having switchable I/O bus interface modules
WO1997007457A1 (en) * 1995-08-14 1997-02-27 Data General Corporation A high availability computer system and methods related thereto
US5692121A (en) * 1995-04-14 1997-11-25 International Business Machines Corporation Recovery unit for mirrored processors
US5812822A (en) * 1995-12-19 1998-09-22 Selway; David W. Apparatus for coordinating clock oscillators in a fully redundant computer system
US6032265A (en) * 1995-07-18 2000-02-29 Hitachi, Ltd. Fault-tolerant computer system
US20020124202A1 (en) * 2001-03-05 2002-09-05 John Doody Coordinated Recalibration of high bandwidth memories in a multiprocessor computer
US6597204B2 (en) * 2000-11-10 2003-07-22 Nec Corporation Clock interruption detection circuit
US6718474B1 (en) 2000-09-21 2004-04-06 Stratus Technologies Bermuda Ltd. Methods and apparatus for clock management based on environmental conditions
US6766413B2 (en) 2001-03-01 2004-07-20 Stratus Technologies Bermuda Ltd. Systems and methods for caching with file-level granularity
US6832347B1 (en) * 1999-06-08 2004-12-14 Cisco Technology, Inc. Clock synchronization and fault protection for a telecommunications device
US6862689B2 (en) 2001-04-12 2005-03-01 Stratus Technologies Bermuda Ltd. Method and apparatus for managing session information
US20060222126A1 (en) * 2005-03-31 2006-10-05 Stratus Technologies Bermuda Ltd. Systems and methods for maintaining synchronicity during signal transmission
US20060222125A1 (en) * 2005-03-31 2006-10-05 Edwards John W Jr Systems and methods for maintaining synchronicity during signal transmission
US7350116B1 (en) 1999-06-08 2008-03-25 Cisco Technology, Inc. Clock synchronization and fault protection for a telecommunications device

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2311881B (en) * 1996-04-03 2000-03-29 Ind Control Services Technolog Fault tolerant data processing systems
FR2818871B1 (en) * 2000-12-21 2003-03-07 Crouzet Automatismes control system of a switch has two states

Citations (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE2360450A1 (en) * 1973-12-01 1975-06-05 Licentia Gmbh Fail-safe three-channel clock generator - remains operational even if one of three channel pulse generators fails
US3900741A (en) * 1973-04-26 1975-08-19 Nasa Fault tolerant clock apparatus utilizing a controlled minority of clock elements
US4019143A (en) * 1976-05-10 1977-04-19 Bell Telephone Laboratories, Incorporated Standby apparatus for clock signal generators
US4025874A (en) * 1976-04-30 1977-05-24 Rockwell International Corporation Master/slave clock arrangement for providing reliable clock signal
US4144448A (en) * 1977-11-29 1979-03-13 International Business Machines Corporation Asynchronous validity checking system and method for monitoring clock signals on separate electrical conductors
US4156200A (en) * 1978-03-20 1979-05-22 Bell Telephone Laboratories, Incorporated High reliability active-standby clock arrangement
US4185245A (en) * 1978-05-15 1980-01-22 International Telephone And Telegraph Corporation Fault-tolerant clock signal distribution arrangement
US4239982A (en) * 1978-06-14 1980-12-16 The Charles Stark Draper Laboratory, Inc. Fault-tolerant clock system
US4322580A (en) * 1980-09-02 1982-03-30 Gte Automatic Electric Labs Inc. Clock selection circuit
JPS5773532A (en) * 1980-10-24 1982-05-08 Mitsubishi Electric Corp Count value storage device
US4453215A (en) * 1981-10-01 1984-06-05 Stratus Computer, Inc. Central processing apparatus for fault-tolerant computing
US4480198A (en) * 1981-05-20 1984-10-30 La Telephonie Industrielle Et Commerciale Telic Alcatel Device for increasing the operational security of a duplicated clock
US4538272A (en) * 1983-12-22 1985-08-27 Gte Automatic Electric Incorporated Prioritized clock selection circuit
US4653054A (en) * 1985-04-12 1987-03-24 Itt Corporation Redundant clock combiner
US4686677A (en) * 1985-08-02 1987-08-11 Unisys Corporation Apparatus and method for detecting time-related faults
US4691126A (en) * 1985-08-29 1987-09-01 Sperry Corporation Redundant synchronous clock system
US4800564A (en) * 1986-09-29 1989-01-24 International Business Machines Corporation High performance clock system error detection and fault isolation

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US3628158A (en) * 1968-11-15 1971-12-14 Ericsson Telefon Ab L M Arrangement at parallelly working machines

Patent Citations (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US3900741A (en) * 1973-04-26 1975-08-19 Nasa Fault tolerant clock apparatus utilizing a controlled minority of clock elements
DE2360450A1 (en) * 1973-12-01 1975-06-05 Licentia Gmbh Fail-safe three-channel clock generator - remains operational even if one of three channel pulse generators fails
US4025874A (en) * 1976-04-30 1977-05-24 Rockwell International Corporation Master/slave clock arrangement for providing reliable clock signal
US4019143A (en) * 1976-05-10 1977-04-19 Bell Telephone Laboratories, Incorporated Standby apparatus for clock signal generators
US4144448A (en) * 1977-11-29 1979-03-13 International Business Machines Corporation Asynchronous validity checking system and method for monitoring clock signals on separate electrical conductors
US4156200A (en) * 1978-03-20 1979-05-22 Bell Telephone Laboratories, Incorporated High reliability active-standby clock arrangement
US4185245A (en) * 1978-05-15 1980-01-22 International Telephone And Telegraph Corporation Fault-tolerant clock signal distribution arrangement
US4239982A (en) * 1978-06-14 1980-12-16 The Charles Stark Draper Laboratory, Inc. Fault-tolerant clock system
US4322580A (en) * 1980-09-02 1982-03-30 Gte Automatic Electric Labs Inc. Clock selection circuit
JPS5773532A (en) * 1980-10-24 1982-05-08 Mitsubishi Electric Corp Count value storage device
US4480198A (en) * 1981-05-20 1984-10-30 La Telephonie Industrielle Et Commerciale Telic Alcatel Device for increasing the operational security of a duplicated clock
US4453215A (en) * 1981-10-01 1984-06-05 Stratus Computer, Inc. Central processing apparatus for fault-tolerant computing
US4538272A (en) * 1983-12-22 1985-08-27 Gte Automatic Electric Incorporated Prioritized clock selection circuit
US4653054A (en) * 1985-04-12 1987-03-24 Itt Corporation Redundant clock combiner
US4686677A (en) * 1985-08-02 1987-08-11 Unisys Corporation Apparatus and method for detecting time-related faults
US4691126A (en) * 1985-08-29 1987-09-01 Sperry Corporation Redundant synchronous clock system
US4800564A (en) * 1986-09-29 1989-01-24 International Business Machines Corporation High performance clock system error detection and fault isolation

Non-Patent Citations (18)

* Cited by examiner, † Cited by third party
Title
Hefferman et al., "Missing Clock Pulse Detector", IBM TDB, vol. 5, No. 1, Jun./62, p.76.
Hefferman et al., Missing Clock Pulse Detector , IBM TDB, vol. 5, No. 1, Jun./62, p.76. *
IBM Research, C. D. Cullum, "A Survey of Encoding and Signal Processing Techniques for Digital Magnetic Recording," Mar. 4, 1971.
IBM Research, C. D. Cullum, A Survey of Encoding and Signal Processing Techniques for Digital Magnetic Recording, Mar. 4, 1971. *
IBM Technical Bulletin, "Checking Digital System Clock Pulses", vol. 11, No. 3, Aug./68, pp. 257-258.
IBM Technical Bulletin, "Clock/Synch Failure Detector Apparatus", vol. 22, No. 4, Sep./79, pp. 1495-1496.
IBM Technical Bulletin, "Fail-Safe Clock Generation by Majority Logic", vol. 27, No. 4B, Sep./84, pp. 2640-2641.
IBM Technical Bulletin, "Missing Clock Detector", vol. 19, No. 5, Oct./76, pp. 1905-1906.
IBM Technical Bulletin, "Synchronized Clocking System", vol. 19, No. 5, Oct./76, pp. 1900-1901.
IBM Technical Bulletin, Checking Digital System Clock Pulses , vol. 11, No. 3, Aug./68, pp. 257 258. *
IBM Technical Bulletin, Clock/Synch Failure Detector Apparatus , vol. 22, No. 4, Sep./79, pp. 1495 1496. *
IBM Technical Bulletin, Fail Safe Clock Generation by Majority Logic , vol. 27, No. 4B, Sep./84, pp. 2640 2641. *
IBM Technical Bulletin, Missing Clock Detector , vol. 19, No. 5, Oct./76, pp. 1905 1906. *
IBM Technical Bulletin, Synchronized Clocking System , vol. 19, No. 5, Oct./76, pp. 1900 1901. *
Standard Microsystems Corporation Technical Bulletin, "Floppy Disk Data Separator FDDS," 1982.
Standard Microsystems Corporation Technical Bulletin, Floppy Disk Data Separator FDDS, 1982. *
Sworakowski, "Redundant Clock Magnetic Recording", IBM TDB, vol. 12, No. 2, Jul./69, p. 254.
Sworakowski, Redundant Clock Magnetic Recording , IBM TDB, vol. 12, No. 2, Jul./69, p. 254. *

Cited By (28)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5020024A (en) * 1987-01-16 1991-05-28 Stratus Computer, Inc. Method and apparatus for detecting selected absence of digital logic synchronism
US5588111A (en) * 1988-12-09 1996-12-24 Tandem Computers, Incorporated Fault-tolerant computer system having switchable I/O bus interface modules
US5077739A (en) * 1989-05-17 1991-12-31 Unisys Corporation Method for isolating failures of clear signals in instruction processors
US5220668A (en) * 1990-09-21 1993-06-15 Stratus Computer, Inc. Digital data processor with maintenance and diagnostic system
US5440751A (en) * 1991-06-21 1995-08-08 Compaq Computer Corp. Burst data transfer to single cycle data transfer conversion and strobe signal conversion
US5313476A (en) * 1991-06-28 1994-05-17 International Business Machines Corporation Clock security ring
US5257383A (en) * 1991-08-12 1993-10-26 Stratus Computer, Inc. Programmable interrupt priority encoder method and apparatus
US5379381A (en) * 1991-08-12 1995-01-03 Stratus Computer, Inc. System using separate transfer circuits for performing different transfer operations respectively and scanning I/O devices status upon absence of both operations
US5553231A (en) * 1992-09-29 1996-09-03 Zitel Corporation Fault tolerant memory system
US5379415A (en) * 1992-09-29 1995-01-03 Zitel Corporation Fault tolerant memory system
US5377206A (en) * 1993-02-03 1994-12-27 Honeywell Inc. Multiple-channel fault-tolerant clock system
US5422896A (en) * 1993-02-24 1995-06-06 Nec Corporation Timing check circuit for a functional macro
US5471488A (en) * 1994-04-05 1995-11-28 International Business Machines Corporation Clock fault detection circuit
US5692121A (en) * 1995-04-14 1997-11-25 International Business Machines Corporation Recovery unit for mirrored processors
US6032265A (en) * 1995-07-18 2000-02-29 Hitachi, Ltd. Fault-tolerant computer system
WO1997007457A1 (en) * 1995-08-14 1997-02-27 Data General Corporation A high availability computer system and methods related thereto
US6122756A (en) * 1995-08-14 2000-09-19 Data General Corporation High availability computer system and methods related thereto
US5812822A (en) * 1995-12-19 1998-09-22 Selway; David W. Apparatus for coordinating clock oscillators in a fully redundant computer system
US6832347B1 (en) * 1999-06-08 2004-12-14 Cisco Technology, Inc. Clock synchronization and fault protection for a telecommunications device
US7350116B1 (en) 1999-06-08 2008-03-25 Cisco Technology, Inc. Clock synchronization and fault protection for a telecommunications device
US6718474B1 (en) 2000-09-21 2004-04-06 Stratus Technologies Bermuda Ltd. Methods and apparatus for clock management based on environmental conditions
US6597204B2 (en) * 2000-11-10 2003-07-22 Nec Corporation Clock interruption detection circuit
US6766413B2 (en) 2001-03-01 2004-07-20 Stratus Technologies Bermuda Ltd. Systems and methods for caching with file-level granularity
US20020124202A1 (en) * 2001-03-05 2002-09-05 John Doody Coordinated Recalibration of high bandwidth memories in a multiprocessor computer
US6874102B2 (en) 2001-03-05 2005-03-29 Stratus Technologies Bermuda Ltd. Coordinated recalibration of high bandwidth memories in a multiprocessor computer
US6862689B2 (en) 2001-04-12 2005-03-01 Stratus Technologies Bermuda Ltd. Method and apparatus for managing session information
US20060222126A1 (en) * 2005-03-31 2006-10-05 Stratus Technologies Bermuda Ltd. Systems and methods for maintaining synchronicity during signal transmission
US20060222125A1 (en) * 2005-03-31 2006-10-05 Edwards John W Jr Systems and methods for maintaining synchronicity during signal transmission

Also Published As

Publication number Publication date Type
DE3853734T2 (en) 1996-02-01 grant
JPS63296118A (en) 1988-12-02 application
EP0280258A3 (en) 1990-05-02 application
EP0280258A2 (en) 1988-08-31 application
DE3853734D1 (en) 1995-06-14 grant
EP0280258B1 (en) 1995-05-10 grant

Similar Documents

Publication Publication Date Title
US5101117A (en) Variable delay line phase-locked loop circuit synchronization system
US4973860A (en) Circuit for synchronizing an asynchronous input signal to a high frequency clock
US4674036A (en) Duplex controller synchronization circuit for processors which utilizes an address input
US5422915A (en) Fault tolerant clock distribution system
US6069506A (en) Method and apparatus for improving the performance of digital delay locked loop circuits
US4868514A (en) Apparatus and method for digital compensation of oscillator drift
US6848060B2 (en) Synchronous to asynchronous to synchronous interface
US5675274A (en) Semiconductor clock signal generation circuit
US5274678A (en) Clock switching apparatus and method for computer systems
US5987081A (en) Method and apparatus for a testable high frequency synchronizer
US5808571A (en) Synchronization control unit which maintains synchronization between serial-to-parallel converters operating in parallel, or between parallel-to-serial converters operating in parallel
US3479603A (en) A plurality of sources connected in parallel to produce a timing pulse output while any source is operative
US6107841A (en) Synchronous clock switching circuit for multiple asynchronous clock source
US6516422B1 (en) Computer system including multiple clock sources and failover switching
US5425074A (en) Fast programmable/resettable CMOS Johnson counters
US6208180B1 (en) Core clock correction in a 2/N mode clocking scheme
US5517147A (en) Multiple-phase clock signal generator for integrated circuits, comprising PLL, counter, and logic circuits
US5394443A (en) Multiple interval single phase clock
US5900757A (en) Clock stopping schemes for data buffer
US5811998A (en) State machine phase lock loop
US5535377A (en) Method and apparatus for low latency synchronization of signals having different clock speeds
US6247137B1 (en) Delaying clock and data signals to force synchronous operation in digital systems that determine phase relationships between clocks with related frequencies
US6622256B1 (en) System for protecting strobe glitches by separating a strobe signal into pointer path and timing path, filtering glitches from signals on pointer path thereof
US5345109A (en) Programmable clock circuit
US5623223A (en) Glitchless clock switching circuit

Legal Events

Date Code Title Description
AS Assignment

Owner name: STRATUS COMPUTER, INC., 55 FAIRBANKS BLVD., MARLBO

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST.;ASSIGNOR:BATY, KURT F.;REEL/FRAME:004696/0605

Effective date: 19870220

FPAY Fee payment

Year of fee payment: 4

FPAY Fee payment

Year of fee payment: 8

LAPS Lapse for failure to pay maintenance fees
FP Expired due to failure to pay maintenance fee

Effective date: 20020424